Re: Help! RDF & HTML Compatibility

HTML pages CAN reference RDF datra externally.  The W3C Note "Accessibility
Features of SVG" - - does just that. And the
idea of namespaces is that this need not be the case for ever. That kind of
technology is still in early stage development - Amaya uses it in a rather
specialised way to mix MathML, SVG and XHTML, and a number of RDF tools are
happy to deal with it.  But as David points out, there is not a monstrous
range of general purpose applications trying it - Office 2000 is the only one
that springs to mind.

HTML is of course a special case, since prior to XHTML it wasn't an XML
application anyway. The processing model of HTML was a bit messy, and the
implementations much more so. I think one of the goals of XHTML was to
provide HTML with a path to the more reliable parsing the XML world uses.
Coping with the state of the world as it was left us with two different RDF
syntaxes. There are several ways of mixing XML types (maybe that means it is
a common desire), including DTD modularisation, schemas, namespaces, and use
of external references. Each of them has particular strengths and weaknesses.
But I am not sure that we should rule out forever the idea of mixing content
types. (A more extreme example is adding RDF to jpg images - which seems a
particularly useful thing to do...)

charles McCN

On Sat, 23 Dec 2000, David Megginson wrote:

  Dan Brickley writes:

   > Yes, the love-it-or-loathe-it syntactic flexibility of RDF stems in
   > large part from this issue. We needed to have a way of writing RDF
   > assertions within documents that would be fed to old style HTML
   > rendering engines, and not have chunks of RDF data spew out into the
   > human oriented view of the document.

  I'll add the footnote that a group of us on the XML WG argued
  vigorously that HTML pages should reference RDF and similar XML-based
  data externally, as they do with images, Java applets, and sound clips
  (to name only three).  That would have saved all of the ugly syntactic
  contortions for RDF and P3P (which used RDF at the time), and would
  have allowed us to avoid using attributes for Namespace declarations.

  We lost the debate, of course.  Two years later, it's hard to know who
  was right, or whether it even mattered -- client-side XML on the Web
  remains virtually non-existant, and nearly all paying XML work is
  business-to-business (even with RDF).

  All the best,


Charles McCathieNevile    phone: +61 (0) 409 134 136
W3C Web Accessibility Initiative            
Location: I-cubed, 110 Victoria Street, Carlton VIC 3053, Australia
until 6 January 2001 at:
W3C INRIA, 2004 Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France

Received on Saturday, 23 December 2000 14:27:44 UTC